Portal de Periódicos da CAPES

Girlboss? Highlighting versus downplaying gender through language

2022; Elsevier BV; Volume: 26; Issue: 8 Linguagem: Inglês

10.1016/j.tics.2022.05.001

1879-307X

Stav Atir,

Multilingual Education and Policy

Emerging research points to the power of language to shape how we think of gender in the professional domain. However, there is tension between two opposing strategies for communicating gender: gender marking and gender neutrality. Each strategy has the potential to combat gender bias, but also to reinforce it. Emerging research points to the power of language to shape how we think of gender in the professional domain. However, there is tension between two opposing strategies for communicating gender: gender marking and gender neutrality. Each strategy has the potential to combat gender bias, but also to reinforce it. Gender is perhaps the demographic identity most often and easily communicated through language. When it comes to speaking about professionals, emerging research suggests that even subtle linguistic differences in word choice or phrasing can reduce or exacerbate gender bias. However, the practical takeaways are less clear. Specifically, there is tension in the literature between two types of linguistic practices: gender marking, which emphasizes an individual's gender, and gender neutrality, which de-emphasizes it. Though both strategies have drawbacks, both have strengths as well. By understanding their costs and benefits, individuals can tailor their language to their situation-specific goals. Importantly, I draw here largely on research that assumes binary gender. We are only beginning to learn how language interacts with nonbinary conceptualizations of gender (see Box 1).Box 1Language and nonbinary genderLanguage influences not only perceptions of professionals of different genders, but also perceptions of gender as a concept. Having just two basic gender categories – man/woman or male/female – dichotomizes gender, which is becoming increasingly relevant as more people recognize that not everyone fits into neat, distinct, gender boxes. Though little research has examined how language interacts with nonbinary conceptualizations of gender, we can speculate that gender marking and gender neutrality would pose problems. Gender marking, at least as currently used, requires choosing a specific gender category (e.g., businessman or businesswoman). But for some, neither is a good fit. Gender marking also often requires judging gender based on indirect cues such as appearance, potentially leading to misgendering. The gender-neutral businessperson is technically inclusive, but the same male-default thinking that makes women disappear in gender-neutral language likely makes nonbinary people disappear, too. Even writing about gender in a nonbinary way is difficult using existing linguistic tools. Depending on the audience, 'woman' can mean 'person who identifies as a woman,' 'person whom society largely recognizes as a woman,' or even 'person whose sex is female.' We need better language to capture the nuances of gender identity. Language influences not only perceptions of professionals of different genders, but also perceptions of gender as a concept. Having just two basic gender categories – man/woman or male/female – dichotomizes gender, which is becoming increasingly relevant as more people recognize that not everyone fits into neat, distinct, gender boxes. Though little research has examined how language interacts with nonbinary conceptualizations of gender, we can speculate that gender marking and gender neutrality would pose problems. Gender marking, at least as currently used, requires choosing a specific gender category (e.g., businessman or businesswoman). But for some, neither is a good fit. Gender marking also often requires judging gender based on indirect cues such as appearance, potentially leading to misgendering. The gender-neutral businessperson is technically inclusive, but the same male-default thinking that makes women disappear in gender-neutral language likely makes nonbinary people disappear, too. Even writing about gender in a nonbinary way is difficult using existing linguistic tools. Depending on the audience, 'woman' can mean 'person who identifies as a woman,' 'person whom society largely recognizes as a woman,' or even 'person whose sex is female.' We need better language to capture the nuances of gender identity. Gender marking has clear benefits. In order to spotlight the breakers of glass ceilings and those following in their footsteps, we must mention their gender. If we discuss the CEO of YouTube, for instance, or the 2020 Nobel Laureate in Physics, and don't mention they are women (Susan Wojcicki and Andrea Ghez, respectively), we'd be missing an opportunity to change people's perceptions of who belongs and who can be successful in these professions [1.Kray L.J. et al.The effects of implicit gender role theories on gender system justification: Fixed beliefs strengthen masculinity to preserve the status quo.J. Pers. Soc. Psychol. 2017; 112: 98Crossref PubMed Scopus (54) Google Scholar]. However, gender marking through language looks less appealing when we consider its implementation. How do we signal that we are talking about a woman, for instance? We could use gendered occupation words, such as saying 'businesswoman' or 'businessman,' depending on gender. One issue is that there are far more businessmen than businesswomen, so people would hear 'businessman' far more often than 'businesswoman.' Such gendered language would therefore reflect the skewed gender composition of most high-status professions rather than help undo it. There is also reason to believe that gendered occupation words reinforce gender stereotypes: On the interlanguage level, languages that use more gendered occupation words (e.g., seamstress vs. tailor) have speakers with less egalitarian gender beliefs [2.Lewis M. Lupyan G. Gender stereotypes are reflected in the distributional structure of 25 languages.Nat. Hum. Behav. 2020; 4: 1021-1028Crossref PubMed Scopus (39) Google Scholar]. Although the causal connection is unclear, it's possible that marking occupations by gender perpetuates gender stereotypes by making male and female professionals seem fundamentally different. Another approach to gender marking is to mark gender only when it is atypical. For instance, we could mention a CEO's gender if she's a woman but not if he's a man, or a nurse's gender if he is a man but not if she is a woman. This may indeed make these counterstereotypical exemplars more visible, but it would also spotlight their atypicality. If anyone suggested saying 'female politician' or 'lady scientist,' I think many would say No, thank you. This is because we intuitively understand that using a different word for women in male-dominated fields suggests that these women are aberrations – exceptions that prove the rule. The truth is that people already mark gender-atypical individuals and behaviors in this way. An extreme example is coining slang words for people in gender-atypical occupations, such as 'SheEO' for a female CEO and 'manny' for a male nanny, despite the original occupation words not being definitionally gendered. Another common example is the use of gender marking in sports; we watch 'basketball' if the players are men but 'women's basketball' if they are women [3.Messner M.A. et al.Separating the men from the girls: The gendered language of televised sports.Gend. Soc. 1993; 7: 121-137Crossref Scopus (156) Google Scholar]. People also engage in subtler gender marking, for instance using professionals' names differently depending on these professionals' gender. Specifically, in many professional fields, it is common to refer to others using their surname when talking about them or their work; for example, in sports, 'Federer and Nadal are top tennis players'; in science, 'Darwin's theory of evolution'; in politics, 'Obama's health plan'; in literature, 'Dickens was a prolific author.' Using this surname-only reference does not specify the professional's gender, whereas including a first name often implies the gender. Research shows that people are less likely to use a surname-only reference when talking about female professionals than male professionals, instead using forms of reference that are more likely to reveal gender: full name or first name [4.Atir S. Ferguson M.J. How gender determines the way we speak about professionals.Proc. Natl. Acad. Sci. U. S. A. 2018; 115: 7278-7283Crossref PubMed Scopus (21) Google Scholar]. It may seem cost-free that people are more likely to say 'Pence' than 'Harris,' but research suggests that using a surname-only reference increases perceptions of the professional's fame, eminence, and even deservingness of awards – judgments that people already underattribute to women. Gender marking, then, should not be used thoughtlessly. Though it can draw attention to professionals whose gender is underrepresented, it can also have ironic consequences, prompting stereotypical thinking and bolstering the perception of women as exotic exceptions to the male rule. Instead of making gender salient in the professional sphere, a different linguistic strategy is to minimize the importance of gender through gender neutrality. Gender neutrality of the nonlinguistic variety has famously worked well in orchestra auditions [5.Goldin C. Rouse C. Orchestrating impartiality: The impact of' blind' auditions on female musicians.Am. Econ. Rev. 2000; 90: 715-741Crossref Scopus (760) Google Scholar]: A widely held belief that women are less musically gifted than men led to disparities in the gender make-up of orchestras, until audition practices changed to have candidates play from behind a curtain. The hiring of women to orchestras skyrocketed. Concealing candidates' gender helped curb prejudiced hiring decisions. In similar fashion, we may opt for language that hides gender in the professional sphere as a way of de-emphasizing gender's role. This approach is appealing because in many cases, one's gender is largely irrelevant to one's ability to do the job. One problem with gender-neutral language, however, is that it may not be gender-neutral after all. Even when gender isn't explicitly specified, stereotypes often fill in the gender blank [6.Belle D. et al.'I can't operate, that boy is my son!': Gender schemas and a classic riddle.Sex Roles. 2021; 85: 161-171Crossref Scopus (4) Google Scholar]. Occupation words such as 'businessperson' or 'surgeon,' though technically gender-neutral, likely conjure up an image of a man; likewise, 'nurse' (also technically gender-neutral) conjures up an image of a woman. Using so-called gender-neutral occupation words affects not only our resulting mental image, but also whether we think we fit that image or not; when girls hear 'let's be scientists,' they are less interested in engaging in a science activity than when they hear 'let's do science' [7.Rhodes M. et al.Subtle linguistic cues increase girls' engagement in science.Psychol. Sci. 2019; 30: 455-466Crossref PubMed Scopus (33) Google Scholar], presumably because 'scientist' is a male-stereotyped occupation word that makes girls think the activity is not for them. Gender also permeates verbs and adjectives that are, strictly speaking, gender-neutral [8.Bolukbasi T. et al.Man is to computer programmer as woman is to homemaker? Debiasing word embeddings.Adv. Neural Inf. Proces. Syst. 2016; 29: 1-9Google Scholar, 9.Bian L. et al.Gender stereotypes about intellectual ability emerge early and influence children's interests.Science. 2017; 355: 389-391Crossref PubMed Scopus (323) Google Scholar, 10.Gaucher D. et al.Evidence that gendered wording in job advertisements exists and sustains gender inequality.J. Pers. Soc. Psychol. 2011; 101: 109Crossref PubMed Scopus (242) Google Scholar]. Using new text analysis techniques, linguists find that certain ability adjectives, such as 'genius' and 'brilliant,' are associatively closer to 'he' than to 'she,' with the reverse being true for the occupation word 'homemaker' [8.Bolukbasi T. et al.Man is to computer programmer as woman is to homemaker? Debiasing word embeddings.Adv. Neural Inf. Proces. Syst. 2016; 29: 1-9Google Scholar]. Another issue for gender-neutral language is the 'masculine default.' Even in the absence of relevant stereotypes, as in the sentence, 'A person was walking down the street,' people tend to assume a man by default [11.Bailey A.H. et al.Is man the measure of all things? A social cognitive account of androcentrism.Personal. Soc. Psychol. Rev. 2019; 23: 307-331Crossref PubMed Scopus (49) Google Scholar]. Indeed, according to English dictionaries, 'he' can be used in a generic sense or when gender is unspecified. Yet, mountains of evidence demonstrate that a sentence such as, 'An engineer must plan before he acts,' strongly brings to mind a male engineer and interferes with the mental accessibility of female engineers (see summary of research in [12.Sczesny S. et al.Can gender-fair language reduce gender stereotyping and discrimination?.Front. Psychol. 2016; 7: 25Crossref PubMed Scopus (91) Google Scholar]). Replacing 'he' with the gender-neutral 'they' is only slightly effective for increasing the mental accessibility of female professionals because male is the default anyway. The biggest gains come from explicitly mentioning women, as in 'he or she' or 'congressman or congresswoman' (e.g., [13.Vervecken D. et al.Changing (s)expectations: How gender fair job descriptions impact children's perceptions and interest regarding traditionally male occupations.J. Vocat. Behav. 2013; 82: 208-220Crossref Scopus (61) Google Scholar]). This slight tweak not only makes women professionals more cognitively accessible, but also makes kids less male-biased in who they think could succeed in male-dominated fields, and even leads girls to express more interest in these fields [13.Vervecken D. et al.Changing (s)expectations: How gender fair job descriptions impact children's perceptions and interest regarding traditionally male occupations.J. Vocat. Behav. 2013; 82: 208-220Crossref Scopus (61) Google Scholar]. A well-intentioned wish to de-emphasize gender via gender-neutral language can result in language that conceals women's representation and achievements. Using gendered language such as 'girlboss' and 'guyliner' risks reinforcing the myth that women and men are wildly different creatures, suited to different jobs. It can exacerbate the stereotypical thinking that comes so naturally to us, leading people to overestimate the real differences between genders [14.Eyal T. Epley N. Exaggerating accessible differences: When gender stereotypes overestimate actual group differences.Personal. Soc. Psychol. Bull. 2017; 43: 1323-1336Crossref PubMed Scopus (8) Google Scholar,15.Hyde J.S. The gender similarities hypothesis.Am. Pyschol. 2005; 60: 581-592Crossref PubMed Scopus (1778) Google Scholar]. But wholesale gender neutrality in language is no panacea. Gender neutrality can make women invisible, especially in male-dominated fields, and leave people's stereotypes intact. Though gender is the demographic most readily communicated via language, the same tension exists for other identities as well. For instance, the Western default is not only male, but White, making race-neutral language problematic – but highlighting the race of individuals from under-represented identities risks stereotyping them. There may also be interactions between gender and race, with gendered language affecting women of different races differently. We might be tempted to throw up our hands and give up the endeavor of using language to express and promote our beliefs. That would be a mistake. Language remains one tool in our toolbox for social change, and, unlike some of our other tools, it's one that we can all use. The key to using this tool effectively is to tailor our language to the context, taking into account our situation-specific goals. It's important to remember that gender marking allows us to spotlight the impressive achievements of women in male- and female-dominated fields and that gender neutrality helps us signal, 'I don't think gender matters here,' as well as, 'This occupation isn't only for people of one gender.' Language is nuanced, and our linguistic choices should be as well. I thank Melissa Ferguson and Jane Risen for valuable feedback on this piece. The author has no interests to declare.